Skip to contents

glyvis

The goal of glyvis is to visualize everything in the glycoverse ecosystem. Visualization is an essential part of data analysis. Human beings are more sensitive to visual information than text and numbers. Plotting helps us to understand the data better. glyvis provides a unified interface for visualizing glycoverse data, including statistical results, experiments, glycan biosynthesis pathways, and more. It implements the autoplot() method for various glycoverse data structures. Just autoplot() it!

Installation

Install glycoverse

We recommend installing the meta-package glycoverse, which includes this package and other core glycoverse packages.

Install glyvis alone

If you don’t want to install all glycoverse packages, you can only install glyvis.

You can install the latest release of glyvis from r-universe (recommended):

# install.packages("pak")
pak::repo_add(glycoverse = "https://glycoverse.r-universe.dev")
pak::pkg_install("glyvis")

Or from GitHub:

pak::pkg_install("glycoverse/glyvis@*release")

Or install the development version (NOT recommended):

pak::pkg_install("glycoverse/glyvis")

Note: Tips and troubleshooting for the meta-package glycoverse are also applicable here: Installation of glycoverse.

Documentation

  • πŸš€ Get started: Here
  • πŸ“š Reference: Here

Role in glycoverse

The main purpose of glyvis is to provide visualization for glystats results. It implements autoplot() methods for each result class in glystats, so that the users can visualize the results directly to get a quick overview. It also provides some other visualization functions for glycoverse data structures, such as glyexp::experiment(), glyrepr::glycan_structure(), and others. This package is not intended to produce publication-quality figures, but to provide a quick exploration of the data.

Example 

library(glyexp)
library(glyclean)
#> 
#> θ½½ε…₯η¨‹εΊεŒ…οΌš'glyclean'
#> The following object is masked from 'package:stats':
#> 
#>     aggregate
library(glystats)
library(glyvis)

exp <- auto_clean(real_experiment)
#> 
#> ── Normalizing data ──
#> 
#> β„Ή No QC samples found. Using default normalization method based on experiment type.
#> β„Ή Experiment type is "glycoproteomics". Using `normalize_median()`.
#> βœ” Normalization completed.
#> 
#> ── Removing variables with too many missing values ──
#> 
#> β„Ή No QC samples found. Using all samples.
#> β„Ή Applying preset "discovery"...
#> β„Ή Total removed: 24 (0.56%) variables.
#> βœ” Variable removal completed.
#> 
#> ── Imputing missing values ──
#> 
#> β„Ή No QC samples found. Using default imputation method based on sample size.
#> β„Ή Sample size <= 30, using `impute_sample_min()`.
#> βœ” Imputation completed.
#> 
#> ── Aggregating data ──
#> 
#> β„Ή Aggregating to "gfs" level
#> βœ” Aggregation completed.
#> 
#> ── Normalizing data again ──
#> 
#> β„Ή No QC samples found. Using default normalization method based on experiment type.
#> β„Ή Experiment type is "glycoproteomics". Using `normalize_median()`.
#> βœ” Normalization completed.
#> 
#> ── Correcting batch effects ──
#> 
#> β„Ή Batch column  not found in sample_info. Skipping batch correction.
#> βœ” Batch correction completed.

pca_res <- gly_pca(exp)
autoplot(pca_res)